Dataset Information

Supervised, Self-Administered Tablet-Based Cognitive Assessment in Neurodegenerative Disorders and Stroke.

ABSTRACT:

Introduction

As the population ages, the prevalence of cognitive impairment is expanding. Given the recent pandemic, there is a need for remote testing modalities to assess cognitive deficits in individuals with neurological disorders. Self-administered, remote, tablet-based cognitive assessments would be clinically valuable if they can detect and classify cognitive deficits as effectively as traditional in-person neuropsychological testing.

Methods

We tested whether the Miro application, a tablet-based neurocognitive platform, measured the same cognitive domains as traditional pencil-and-paper neuropsychological tests. Seventy-nine patients were recruited and then randomized to either undergo pencil-and-paper or tablet testing first. Twenty-nine age-matched healthy controls completed the tablet-based assessments. We identified Pearson correlations between Miro tablet-based modules and corresponding neuropsychological tests in patients and compared scores of patients with neurological disorders with those of healthy controls using t tests.

Results

Statistically significant Pearson correlations between the neuropsychological tests and their tablet equivalents were found for all domains with moderate (r > 0.3) or strong (r > 0.7) correlations in 16 of 17 tests (p < 0.05). All tablet-based subtests differentiated healthy controls from neurologically impaired patients by t tests except for the spatial span forward and finger tapping modules. Participants reported enjoyment of the tablet-based testing, denied that it provoked anxiety, and noted no preference between modalities.

Conclusions

This tablet-based application was found to be widely acceptable to participants. This study supports the validity of these tablet-based assessments in the differentiation of healthy controls from patients with neurocognitive deficits in a variety of cognitive domains and across multiple neurological disease etiologies.

SUBMITTER: Sloane KL

PROVIDER: S-EPMC10247386 | biostudies-literature | 2023

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Supervised, Self-Administered Tablet-Based Cognitive Assessment in Neurodegenerative Disorders and Stroke.

Sloane Kelly L KL Fabian Rachel R Wright Amy A Saxena Sadhvi S Kim Kevin K Stein Colin M CM Keser Zafer Z Glenn Shenly S Hillis Argye E AE

Dementia and geriatric cognitive disorders 20230330 2

<h4>Introduction</h4>As the population ages, the prevalence of cognitive impairment is expanding. Given the recent pandemic, there is a need for remote testing modalities to assess cognitive deficits in individuals with neurological disorders. Self-administered, remote, tablet-based cognitive assessments would be clinically valuable if they can detect and classify cognitive deficits as effectively as traditional in-person neuropsychological testing.<h4>Methods</h4>We tested whether the Miro appl ...[more]

PMID: 36996783

Similar Datasets

Project description:BACKGROUND:There is evidence to suggest that rates of hearing loss are increasing more rapidly than the capacity of traditional audiometry resources for screening. A novel innovation in tablet, self-administered portable audiometry has been proposed as a solution to this discordance. The primary objective of this study was to validate a tablet audiometer with adult patients in a clinical setting. Secondarily, word recognition with a tablet audiometer was compared against conventional audiometry. METHODS:Three distinct prospective adult cohorts underwent testing. In group 1 and group 2 testing with the automated tablet audiometer was compared to standard sound booth audiometry. In Group 1, participants' pure tone thresholds were measured with an automated tablet audiometer in a quiet clinic exam room. In Group 2, participants completed monosyllabic word recognition testing using the NU-6 word lists. In Group 3, internal reliability was tested by having participants perform two automated tablet audiometric evaluation in sequence. RESULTS:Group 1 included 40 patients mean age was 54.7 ± 18.4 years old and 60% female; Group 2 included 44 participants mean age was 55.2 ± 14.8 years old and 68.2% female; Group 3 included 40 participants with mean age of 39.4 + 15.9 years old and 60.5% female. In Group 1, compared to standard audiometry, 95.7% (95% CI: 92.6-98.9%) of thresholds were within 10 dB. In Group 2, comparing word recognition results, 96.2% (95% CI: 89.5-98.7%) were clinically equivalent and within a critical difference range. In Group 3, One-way Intraclass Correlation for agreement for the both left- and right-ear pure tone average was 0.98. The mean difference between repeat assessments was 0 (SD = 2.1) in the left ear, and 0.1 (SD = 1.1) in the right ear. CONCLUSION:Puretone audiometry and word recognition testing appears valid when performed by non-healthcare experts using a tablet audiometer outside a sound booth in a quiet environment. TRIAL REGISTRATION:ClinicalTrials.gov Identifier: NCT02761798. Registered April, 2016 < https://clinicaltrials.gov/ct2/show/NCT02761798>.

Project description:IntroductionNeurodegenerative diseases such as Alzheimer's disease (AD) or frontotemporal lobar degeneration (FTLD) involve specific loss of brain volume, detectable in vivo using T1-weighted MRI scans. Supervised machine learning approaches classifying neurodegenerative diseases require diagnostic-labels for each sample. However, it can be difficult to obtain expert labels for a large amount of data. Self-supervised learning (SSL) offers an alternative for training machine learning models without data-labels.MethodsWe investigated if the SSL models can be applied to distinguish between different neurodegenerative disorders in an interpretable manner. Our method comprises a feature extractor and a downstream classification head. A deep convolutional neural network, trained with a contrastive loss, serves as the feature extractor that learns latent representations. The classification head is a single-layer perceptron that is trained to perform diagnostic group separation. We used N = 2,694 T1-weighted MRI scans from four data cohorts: two ADNI datasets, AIBL and FTLDNI, including cognitively normal controls (CN), cases with prodromal and clinical AD, as well as FTLD cases differentiated into its phenotypes.ResultsOur results showed that the feature extractor trained in a self-supervised way provides generalizable and robust representations for the downstream classification. For AD vs. CN, our model achieves 82% balanced accuracy on the test subset and 80% on an independent holdout dataset. Similarly, the Behavioral variant of frontotemporal dementia (BV) vs. CN model attains an 88% balanced accuracy on the test subset. The average feature attribution heatmaps obtained by the Integrated Gradient method highlighted hallmark regions, i.e., temporal gray matter atrophy for AD, and insular atrophy for BV.ConclusionOur models perform comparably to state-of-the-art supervised deep learning approaches. This suggests that the SSL methodology can successfully make use of unannotated neuroimaging datasets as training data while remaining robust and interpretable.

Dataset Information

Supervised, Self-Administered Tablet-Based Cognitive Assessment in Neurodegenerative Disorders and Stroke.

Introduction

Methods

Results

Conclusions

Publications

Supervised, Self-Administered Tablet-Based Cognitive Assessment in Neurodegenerative Disorders and Stroke.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets